Talk given by Francois Meyer of the University of Colorado at Boulder on December 17, 2008 at the Redwood Center for Theoretical Neuroscience at UC Berkeley.
Functional Magnetic Resonance Imaging can be used to infer subjective experience and brain states of subjects immersed in "natural environments". These environments are rich with uncontrolled stimuli and resemble real life experiences. We describe a new method that yields a low dimensional embedding of the fMRI data. The embedding provides a representation of the cognitive activity. We learn a set of time series that are implicit functions of the cognitive activity, and we predict the values of these times series in the future from the knowledge of the fMRI data only. In addition, our exploratory approach is able to detect independently visual areas (V1/V2, V5/MT), auditory areas, and language areas. Our method can be used to analyze fMRI collected during "natural stimuli". We present experiments conducted with the datasets of the 2007 Pittsburgh Brain Activity Interpretation Competition. Reports on this project can be found at: http://ece.colorado.edu/~fmeyer